diff --git a/flepimop/gempyor_pkg/src/gempyor/inference.py b/flepimop/gempyor_pkg/src/gempyor/inference.py
index 823ba52b3..814f687fd 100644
--- a/flepimop/gempyor_pkg/src/gempyor/inference.py
+++ b/flepimop/gempyor_pkg/src/gempyor/inference.py
@@ -454,7 +454,7 @@ def get_all_sim_arguments(self):
     def get_logloss(self, proposal):
         if not self.inferpar.check_in_bound(proposal=proposal):
             if not self.silent:
-                print("`llik` is -inf (out of bound proposal)")
+                print("`llik` is -inf (out of bound proposal).")
             return -np.inf, -np.inf, -np.inf
 
         snpi_df_mod, hnpi_df_mod = self.inferpar.inject_proposal(
diff --git a/flepimop/gempyor_pkg/src/gempyor/model_info.py b/flepimop/gempyor_pkg/src/gempyor/model_info.py
index 35502aadf..3b6cf5afc 100644
--- a/flepimop/gempyor_pkg/src/gempyor/model_info.py
+++ b/flepimop/gempyor_pkg/src/gempyor/model_info.py
@@ -19,7 +19,7 @@ def __init__(self, config: confuse.ConfigView):
         self.tf = config["end_date"].as_date()
         if self.tf <= self.ti:
             raise ValueError(
-                "tf (time to finish) is less than or equal to ti (time to start)"
+                f"Final time ('{self.tf}') is less than or equal to initial time ('{self.ti}')."
             )
         self.n_days = (self.tf - self.ti).days + 1
         self.dates = pd.date_range(start=self.ti, end=self.tf, freq="D")
@@ -76,9 +76,7 @@ def __init__(
         # Auto-detect old config
         if config["interventions"].exists():
             raise ValueError(
-                """This config has an intervention section, and has been written for a previous version of flepiMoP/COVIDScenarioPipeline \
-                             Please use flepiMoP Version 1.1 (Commit SHA: 0c30c23937dd496d33c2b9fa7c6edb198ad80dac) to run this config. \
-                             (use git checkout v1.1 inside the flepiMoP directory)"""
+                "This config has an intervention section, which is only compatible with a previous version (v1.1) of flepiMoP. "
             )
 
         # 1. Create a setup name that contains every scenario.
@@ -104,7 +102,7 @@ def __init__(
         subpop_config = config["subpop_setup"]
         if "data_path" in config:
             raise ValueError(
-                "The config has a data_path section. This is no longer supported."
+                "The config has a `data_path` section. This is no longer supported."
             )
         self.path_prefix = pathlib.Path(path_prefix)
 
@@ -169,19 +167,19 @@ def __init__(
                     self.seir_modifiers_library = config["seir_modifiers"][
                         "modifiers"
                     ].get()
-                    raise ValueError(
-                        "Not implemented yet"
+                    raise NotImplementedError(
+                        "This feature has not been implemented yet."
                     )  # TODO create a Stacked from all
             elif self.seir_modifiers_scenario is not None:
                 raise ValueError(
-                    "An seir modifiers scenario was provided to ModelInfo but no 'seir_modifiers' sections in config"
+                    "A `seir_modifiers_scenario` argument was provided to `ModelInfo` but there is no `seir_modifiers` section in the config."
                 )
             else:
-                logging.info("Running ModelInfo with seir but without SEIR Modifiers")
+                logging.info("Running `ModelInfo` with seir but without SEIR Modifiers")
 
         elif self.seir_modifiers_scenario is not None:
             raise ValueError(
-                "A seir modifiers scenario was provided to ModelInfo but no 'seir:' sections in config"
+                "A `seir_modifiers_scenario` argument was provided to `ModelInfo` but there is no `seir` section in the config."
             )
         else:
             logging.critical("Running ModelInfo without SEIR")
@@ -202,28 +200,28 @@ def __init__(
                     self.outcome_modifiers_library = config["outcome_modifiers"][
                         "modifiers"
                     ].get()
-                    raise ValueError(
-                        "Not implemented yet"
+                    raise NotImplementedError(
+                        "This feature has not been implemented yet."
                     )  # TODO create a Stacked from all
 
             ## NEED TO IMPLEMENT THIS -- CURRENTLY CANNOT USE outcome modifiers
             elif self.outcome_modifiers_scenario is not None:
                 if config["outcome_modifiers"].exists():
                     raise ValueError(
-                        "An outcome modifiers scenario was provided to ModelInfo but no 'outcome_modifiers' sections in config"
+                        "A `outcome_modifiers_scenario` argument was provided to `ModelInfo` but there is no `outcome_modifiers` section in the config."
                     )
                 else:
                     self.outcome_modifiers_scenario = None
             else:
                 logging.info(
-                    "Running ModelInfo with outcomes but without Outcomes Modifiers"
+                    "Running `ModelInfo` with outcomes but without Outcomes Modifiers"
                 )
         elif self.outcome_modifiers_scenario is not None:
             raise ValueError(
-                "An outcome modifiers scenario was provided to ModelInfo but no 'outcomes:' sections in config"
+                "A `outcome_modifiers_scenario` argument was provided to `ModelInfo` but there is no `outcomes` section in the config."
             )
         else:
-            logging.info("Running ModelInfo without Outcomes")
+            logging.info("Running `ModelInfo` without outcomes.")
 
         # 6. Inputs and outputs
         if in_run_id is None:
diff --git a/flepimop/gempyor_pkg/src/gempyor/outcomes.py b/flepimop/gempyor_pkg/src/gempyor/outcomes.py
index e5f09c1d2..7141d0c58 100644
--- a/flepimop/gempyor_pkg/src/gempyor/outcomes.py
+++ b/flepimop/gempyor_pkg/src/gempyor/outcomes.py
@@ -146,7 +146,7 @@ def read_parameters_from_config(modinf: model_info.ModelInfo):
                 branching_data = pa.parquet.read_table(branching_file).to_pandas()
                 if "relative_probability" not in list(branching_data["quantity"]):
                     raise ValueError(
-                        f"No 'relative_probability' quantity in {branching_file}, therefor making it useless"
+                        f"There is no `relative_probability` quantity in '{branching_file}'."
                     )
 
                 print(
@@ -168,7 +168,7 @@ def read_parameters_from_config(modinf: model_info.ModelInfo):
                     modinf.subpop_struct.subpop_names
                 ):
                     raise ValueError(
-                        f"Places in seir input files does not correspond to subpops in outcome probability file {branching_file}"
+                        f"SEIR input files do not have subpops that match those in outcome probability file '{branching_file}'."
                     )
 
         parameters = {}
@@ -184,7 +184,8 @@ def read_parameters_from_config(modinf: model_info.ModelInfo):
 
                 else:
                     raise ValueError(
-                        f"unsure how to read outcome {new_comp}: not a str, nor an incidence or prevalence: {src_name}"
+                        f"Expected a `str` or `dict` containing `incidence` or `prevalence`. "
+                        f"Instead given '{src_name}' for outcome '{new_comp}'."
                     )
 
                 parameters[new_comp]["probability"] = outcomes_config[new_comp][
@@ -295,7 +296,7 @@ def read_parameters_from_config(modinf: model_info.ModelInfo):
                 parameters[new_comp] = {}
                 parameters[new_comp]["sum"] = outcomes_config[new_comp]["sum"].get()
             else:
-                raise ValueError(f"No 'source' or 'sum' specified for comp {new_comp}")
+                raise ValueError(f"No `source` or `sum` specified for comp '{new_comp}'.")
 
     return parameters
 
@@ -396,7 +397,7 @@ def compute_all_multioutcomes(
                     )
                 else:
                     raise ValueError(
-                        f"Unknown type for seir simulation provided, got f{type(seir_sim)}"
+                        f"Unknown type provided for seir simulation, received '{type(seir_sim)}'."
                     )
                 # we don't keep source in this cases
             else:  # already defined outcomes
@@ -404,7 +405,7 @@ def compute_all_multioutcomes(
                     source_array = all_data[source_name]
                 else:
                     raise ValueError(
-                        f"ERROR with outcome {new_comp}: the specified source {source_name} is not a dictionnary (for seir outcome) nor an existing pre-identified outcomes."
+                        f"Issue with outcome '{new_comp}'; the specified source '{source_name}' is neither a dictionnary (for seir outcome) nor an existing pre-identified outcome."
                     )
 
             if (loaded_values is not None) and (
@@ -590,7 +591,7 @@ def filter_seir_df(diffI, dates, subpops, filters, outcome_name) -> np.ndarray:
         vtype = "prevalence"
     else:
         raise ValueError(
-            f"Cannot distinguish the source of outcome {outcome_name}: it is not another previously defined outcome and there is no 'incidence:' or 'prevalence:'."
+            f"Cannot discern the source of outcome '{outcome_name}'; it is not a previously defined outcome and there is no `incidence` or `prevalence`."
         )
 
     diffI = diffI[diffI["mc_value_type"] == vtype]
@@ -619,7 +620,7 @@ def filter_seir_xr(diffI, dates, subpops, filters, outcome_name) -> np.ndarray:
         vtype = "prevalence"
     else:
         raise ValueError(
-            f"Cannot distinguish the source of outcome {outcome_name}: it is not another previously defined outcome and there is no 'incidence:' or 'prevalence:'."
+            f"Cannot discern the source of outcome '{outcome_name}'; it is not a previously defined outcome and there is no `incidence` or `prevalence`."
         )
     # Filter the data
     filters = filters[vtype]
@@ -669,7 +670,7 @@ def multishiftee(arr, shifts, stoch_delay_flag=True):
     result = np.zeros_like(arr)
 
     if stoch_delay_flag:
-        raise ValueError("NOT SUPPORTED YET")
+        raise NotImplementedError("`stoch_delay_flag` not supported yet.")
         # for i, row in reversed(enumerate(np.rows(arr))):
         #    for j,elem in reversed(enumerate(row)):
         ## This function takes in :
@@ -695,7 +696,7 @@ def multishift(arr, shifts, stoch_delay_flag=True):
     result = np.zeros_like(arr)
 
     if stoch_delay_flag:
-        raise ValueError("NOT SUPPORTED YET")
+        raise NotImplementedError("`stoch_delay_flag` not supported yet.")
         # for i, row in reversed(enumerate(np.rows(arr))):
         #    for j,elem in reversed(enumerate(row)):
         ## This function takes in :
diff --git a/flepimop/gempyor_pkg/src/gempyor/seeding.py b/flepimop/gempyor_pkg/src/gempyor/seeding.py
index 53b81c8dc..a2ae8bb88 100644
--- a/flepimop/gempyor_pkg/src/gempyor/seeding.py
+++ b/flepimop/gempyor_pkg/src/gempyor/seeding.py
@@ -17,9 +17,7 @@
 
 def _DataFrame2NumbaDict(df, amounts, modinf) -> nb.typed.Dict:
     if not df["date"].is_monotonic_increasing:
-        raise ValueError(
-            "_DataFrame2NumbaDict got an unsorted dataframe, exposing itself to non-sense"
-        )
+        raise ValueError("The `df` given is not sorted by the 'date' column.")
 
     cmp_grp_names = [
         col for col in modinf.compartments.compartments.columns if col != "name"
@@ -111,7 +109,7 @@ def get_from_config(self, sim_id: int, modinf) -> nb.typed.Dict:
             dupes = seeding[seeding.duplicated(["subpop", "date"])].index + 1
             if not dupes.empty:
                 raise ValueError(
-                    f"Repeated subpop-date in rows {dupes.tolist()} of seeding::lambda_file."
+                    f"There are repeating subpop-date in rows '{dupes.tolist()}' of `seeding::lambda_file`."
                 )
         elif method == "FolderDraw":
             seeding = pd.read_csv(
@@ -136,7 +134,7 @@ def get_from_config(self, sim_id: int, modinf) -> nb.typed.Dict:
             seeding = pd.DataFrame(columns=["date", "subpop"])
             return _DataFrame2NumbaDict(df=seeding, amounts=[], modinf=modinf)
         else:
-            raise NotImplementedError(f"unknown seeding method [got: {method}]")
+            raise ValueError(f"Unknown seeding method given, '{method}'.")
 
         # Sorting by date is very important here for the seeding format necessary !!!!
         # print(seeding.shape)
@@ -160,8 +158,6 @@ def get_from_config(self, sim_id: int, modinf) -> nb.typed.Dict:
             )
         elif method == "FolderDraw" or method == "FromFile":
             amounts = seeding["amount"]
-        else:
-            raise ValueError(f"Unknown seeding method: {method}")
 
         return _DataFrame2NumbaDict(df=seeding, amounts=amounts, modinf=modinf)
 
diff --git a/flepimop/gempyor_pkg/src/gempyor/seir.py b/flepimop/gempyor_pkg/src/gempyor/seir.py
index 5ea236c98..b5742efad 100644
--- a/flepimop/gempyor_pkg/src/gempyor/seir.py
+++ b/flepimop/gempyor_pkg/src/gempyor/seir.py
@@ -31,7 +31,9 @@ def build_step_source_arg(
             if integration_method == "rk4":
                 integration_method = "rk4.jit"
             if integration_method not in ["rk4.jit", "legacy"]:
-                raise ValueError(f"Unknown integration method {integration_method}.")
+                raise ValueError(
+                    f"Unknown integration method given, '{integration_method}'."
+                )
         if "dt" in modinf.seir_config["integration"].keys():
             dt = float(
                 eval(str(modinf.seir_config["integration"]["dt"].get()))
@@ -148,8 +150,7 @@ def steps_SEIR(
     elif integration_method == "rk4.jit":
         if modinf.stoch_traj_flag == True:
             raise ValueError(
-                f"with method {integration_method}, only deterministic "
-                f"integration is possible (got stoch_straj_flag={modinf.stoch_traj_flag}"
+                f"'{integration_method}' integration method only supports deterministic integration, but `stoch_straj_flag` is '{modinf.stoch_traj_flag}'."
             )
         seir_sim = steps_rk4.rk4_integration(**fnct_args, silent=True)
     else:
@@ -164,8 +165,7 @@ def steps_SEIR(
         ]:
             if modinf.stoch_traj_flag == True:
                 raise ValueError(
-                    f"with method {integration_method}, only deterministic "
-                    f"integration is possible (got stoch_straj_flag={modinf.stoch_traj_flag}"
+                    f"'{integration_method}' integration method only supports deterministic integration, but `stoch_straj_flag` is '{modinf.stoch_traj_flag}'."
                 )
             seir_sim = steps_experimental.ode_integration(
                 **fnct_args, integration_method=integration_method
@@ -187,7 +187,7 @@ def steps_SEIR(
         elif integration_method == "rk4_aot":
             seir_sim = steps_experimental.rk4_aot(**fnct_args)
         else:
-            raise ValueError(f"Unknow integration scheme, got {integration_method}")
+            raise ValueError(f"Unknown integration method given, '{integration_method}'.")
 
     # We return an xarray instead of a ndarray now
     compartment_coords = {}
diff --git a/flepimop/gempyor_pkg/src/gempyor/statistics.py b/flepimop/gempyor_pkg/src/gempyor/statistics.py
index ea10e7aa1..180dbe80e 100644
--- a/flepimop/gempyor_pkg/src/gempyor/statistics.py
+++ b/flepimop/gempyor_pkg/src/gempyor/statistics.py
@@ -74,7 +74,10 @@ def __init__(self, name: str, statistic_config: confuse.ConfigView) -> None:
                 reg_name = reg_config["name"].get()
                 reg_func = getattr(self, f"_{reg_name}_regularize", None)
                 if reg_func is None:
-                    raise ValueError(f"Unsupported regularization: {reg_name}")
+                    raise ValueError(
+                        f"Unsupported regularization [received: '{reg_name}']. "
+                        f"Currently only `forecast` and `allsubpop` are supported."
+                    )
                 self.regularizations.append((reg_func, reg_config.get()))
 
         self.resample = False
@@ -253,7 +256,10 @@ def llik(self, model_data: xr.DataArray, gt_data: xr.DataArray) -> xr.DataArray:
             "absolute_error": lambda x, y: -np.log(np.nansum(np.abs(x - y))),
         }
         if self.dist not in dist_map:
-            raise ValueError(f"Invalid distribution specified: {self.dist}")
+            raise ValueError(
+                f"Invalid distribution specified: '{self.dist}'. "
+                f"Valid distributions: '{dist_map.keys()}'."
+            )
         if self.dist in ["pois", "nbinom"]:
             model_data = model_data.astype(int)
             gt_data = gt_data.astype(int)
@@ -295,11 +301,9 @@ def compute_logloss(
 
         if not model_data.shape == gt_data.shape:
             raise ValueError(
-                (
-                    f"{self.name} Statistic error: data and groundtruth do not have "
-                    f"the same shape: model_data.shape={model_data.shape} != "
-                    f"gt_data.shape={gt_data.shape}"
-                )
+                f"`model_data` and `gt_data` do not have "
+                f"the same shape: `model_data.shape` = '{model_data.shape}' != "
+                f"`gt_data.shape` = '{gt_data.shape}'."
             )
 
         regularization = 0.0
diff --git a/flepimop/gempyor_pkg/src/gempyor/steps_source.py b/flepimop/gempyor_pkg/src/gempyor/steps_source.py
index ea6f7ad21..7da7c393d 100644
--- a/flepimop/gempyor_pkg/src/gempyor/steps_source.py
+++ b/flepimop/gempyor_pkg/src/gempyor/steps_source.py
@@ -278,8 +278,11 @@ def steps_SEIR_nb(
                 print("  ", states_current[comp].max())
 
         if (states_current.min() < 0) or (states_current.max() > 10**10):
-            print((states_current.min() < 0), (states_current.max() > 10**10))
-            raise ValueError(f"Overflow error. Too small ?. Too large ?")
+            raise ValueError(
+                f"State values are outside the valid range. "
+                f"Minimum value: '{states_current.min()}', Maximum value: '{states_current.max()}'. "
+                f"Valid range is between 0 and {10**10}."
+            )
 
     return states, states_daily_incid
 
diff --git a/flepimop/gempyor_pkg/src/gempyor/utils.py b/flepimop/gempyor_pkg/src/gempyor/utils.py
index 377dcdee5..9732d70c2 100644
--- a/flepimop/gempyor_pkg/src/gempyor/utils.py
+++ b/flepimop/gempyor_pkg/src/gempyor/utils.py
@@ -58,7 +58,7 @@ def write_df(
     elif path.suffix == ".parquet":
         return df.to_parquet(path, index=False, engine="pyarrow")
     raise NotImplementedError(
-        f"Invalid extension {path.suffix[1:]}. Must be 'csv' or 'parquet'."
+        f"Invalid extension provided: '.{path.suffix[1:]}'. Supported extensions are `.csv` or `.parquet`."
     )
 
 
@@ -95,7 +95,7 @@ def read_df(
     elif path.suffix == ".parquet":
         return pd.read_parquet(path, engine="pyarrow")
     raise NotImplementedError(
-        f"Invalid extension {path.suffix[1:]}. Must be 'csv' or 'parquet'."
+        f"Invalid extension provided: '.{path.suffix[1:]}'. Supported extensions are `.csv` or `.parquet`."
     )
 
 
@@ -114,7 +114,7 @@ def command_safe_run(
         As a tuple; the return code, the standard output, and standard error from running the command.
 
     Raises:
-        Exception: If fail_on_fail=True and the command fails, an exception will be thrown.
+        RuntimeError: If fail_on_fail=True and the command fails, an error will be thrown.
     """
     import subprocess
     import shlex  # using shlex to split the command because it's not obvious https://docs.python.org/3/library/subprocess.html#subprocess.Popen
@@ -124,8 +124,8 @@ def command_safe_run(
     )
     (stdout, stderr) = sr.communicate()
     if sr.returncode != 0:
-        print(f"{command_name} failed failed with returncode {sr.returncode}")
-        print(f"{command_name}:  {command}")
+        print(f"'{command_name}' failed failed with returncode '{sr.returncode}'")
+        print(f"'{command_name}': '{command}'")
         print("{command_name} command failed with stdout and stderr:")
 
         print("{command_name} stdout >>>>>>")
@@ -136,7 +136,7 @@ def command_safe_run(
         print(stderr.decode())
         print("{command_name} stderr <<<<<<")
         if fail_on_fail:
-            raise Exception(f"{command_name} command failed")
+            raise RuntimeError(f"The '{command_name}' command failed.")
 
     return sr.returncode, stdout, stderr
 
@@ -362,7 +362,7 @@ def as_evaled_expression(self):
         except TypeError as e:
             raise ValueError(e) from e
     else:
-        raise ValueError(f"expected numeric or string expression [got: {value}]")
+        raise ValueError(f"Expected numeric or string expression [received: '{value}'].")
 
 
 def get_truncated_normal(
@@ -484,7 +484,7 @@ def random_distribution_sampler(
     elif distribution == "binomial":
         p = kwargs.get("p")
         if not (0 < p < 1):
-            raise ValueError(f"p value {p} is out of range [0,1]")
+            raise ValueError(f"Invalid `p-value`: '{p}' is out of range [0,1].")
         return functools.partial(np.random.binomial, kwargs.get("n"), p)
     elif distribution == "truncnorm":
         # Truncated normal with mean, sd on interval [a, b]
@@ -497,7 +497,7 @@ def random_distribution_sampler(
     elif distribution == "lognorm":
         # Lognormal distribution with meanlog, sdlog
         return get_log_normal(kwargs.get("meanlog"), kwargs.get("sdlog")).rvs
-    raise NotImplementedError(f"unknown distribution [got: {distribution}]")
+    raise NotImplementedError(f"Unknown distribution [received: '{distribution}'].")
 
 
 @add_method(confuse.ConfigView)
@@ -556,7 +556,7 @@ def as_random_distribution(self):
         elif dist == "binomial":
             p = self["p"].as_evaled_expression()
             if (p < 0) or (p > 1):
-                raise ValueError(f"""p value { p } is out of range [0,1]""")
+                raise ValueError(f"Invalid `p-value`: '{p}' is out of range [0,1].")
                 # if (self["p"] < 0) or (self["p"] > 1):
                 #    raise ValueError(f"""p value { self["p"] } is out of range [0,1]""")
             return functools.partial(
@@ -578,7 +578,7 @@ def as_random_distribution(self):
                 sdlog=self["sdlog"].as_evaled_expression(),
             ).rvs
         else:
-            raise NotImplementedError(f"unknown distribution [got: {dist}]")
+            raise NotImplementedError(f"Unknown distribution [received: '{dist}'].")
     else:
         # we allow a fixed value specified directly:
         return functools.partial(
@@ -716,13 +716,13 @@ def bash(command: str) -> str:
         f"shutil.disk_usage: {total_bytes/ 1000000} Mb total, {used_bytes / 1000000} Mb used, {free_bytes / 1000000} Mb free..."
     )
     print("------------")
-    print(f"df -hT: {bash('df -hT')}")
+    print(f"df -hT: '{bash('df -hT')}'")
     print("------------")
-    print(f"df -i: {bash('df -i')}")
+    print(f"df -i: '{bash('df -i')}'")
     print("------------")
-    print(f"free -h: {bash('free -h')}")
+    print(f"free -h: '{bash('free -h')}'")
     print("------------")
-    print(f"lsblk: {bash('lsblk')}")
+    print(f"lsblk: '{bash('lsblk')}'")
     print("END AWS DIAGNOSIS ================================")
 
 
@@ -1004,8 +1004,8 @@ def download_file_from_s3(name_map: dict[str, str]) -> None:
     except ModuleNotFoundError:
         raise ModuleNotFoundError(
             (
-                "No module named 'boto3', which is required for "
-                "gempyor.utils.download_file_from_s3. Please install the aws target."
+                "No module named `boto3` found, which is required for "
+                "`gempyor.utils.download_file_from_s3`. Please install the aws target."
             )
         )
     s3 = boto3.client("s3")
@@ -1020,10 +1020,9 @@ def download_file_from_s3(name_map: dict[str, str]) -> None:
                 object = s3_uri[len(bucket) + 6 :]
                 s3.download_file(bucket, object, name_map[s3_uri])
             else:
-                raise ValueError(f"Invalid S3 URI format {s3_uri}")
+                raise ValueError(f"Invalid S3 URI format [received: '{s3_uri}'].")
         except ClientError as e:
-            print(f"An error occurred: {e}")
-            print("Could not download file from s3")
+            raise Exception(f"'{e}': could not download filefrom S3.")
 
 
 def move_file_at_local(name_map: dict[str, str]) -> None:
diff --git a/flepimop/gempyor_pkg/tests/parameters/test_parameters_class.py b/flepimop/gempyor_pkg/tests/parameters/test_parameters_class.py
index f82cf2a27..9801a6193 100644
--- a/flepimop/gempyor_pkg/tests/parameters/test_parameters_class.py
+++ b/flepimop/gempyor_pkg/tests/parameters/test_parameters_class.py
@@ -283,7 +283,7 @@ def test_timeseries_parameter_has_insufficient_dates_value_error(
             ValueError,
             match=(
                 f"Issue loading file '{tmp_file}' for parameter 'sigma': "
-                f"Provided file dates span '{timeseries_start_date}( 00\:00\:00)?' to '{timeseries_end_date}( 00\:00\:00)?', "
+                f"Provided file dates span '{timeseries_start_date}( 00:00:00)?' to '{timeseries_end_date}( 00:00:00)?', "
                 f"but the config dates span '{mock_inputs.ti}' to '{mock_inputs.tf}'.$"
             ),
         ):
diff --git a/flepimop/gempyor_pkg/tests/seir/test_model_info.py b/flepimop/gempyor_pkg/tests/seir/test_model_info.py
index 3fbf21a75..b241ba263 100644
--- a/flepimop/gempyor_pkg/tests/seir/test_model_info.py
+++ b/flepimop/gempyor_pkg/tests/seir/test_model_info.py
@@ -4,6 +4,7 @@
 import pandas as pd
 import pytest
 import confuse
+import re
 
 from gempyor.model_info import ModelInfo, subpopulation_structure
 
@@ -44,7 +45,10 @@ def test_ModelInfo_init_tf_is_ahead_of_ti_fail(self):
         config.read(user=False)
         config.set_file(f"{DATA_DIR}/config_test.yml")
         config["start_date"] = "2022-01-02"
-        with pytest.raises(ValueError, match=r"tf.*is less than or equal to ti.*"):
+        with pytest.raises(
+            ValueError,
+            match=re.escape(rf"Final time ('{config['end_date']}') is less than or equal to initial time ('{config['start_date']}')."),
+        ):
             s = ModelInfo(
                 config=config,
                 seir_modifiers_scenario=None,
diff --git a/flepimop/gempyor_pkg/tests/seir/test_seir.py b/flepimop/gempyor_pkg/tests/seir/test_seir.py
index d9d6d696f..eeb04fff8 100644
--- a/flepimop/gempyor_pkg/tests/seir/test_seir.py
+++ b/flepimop/gempyor_pkg/tests/seir/test_seir.py
@@ -128,7 +128,9 @@ def test_constant_population_legacy_integration():
 
 
 def test_constant_population_rk4jit_integration_fail():
-    with pytest.raises(ValueError, match=r".*with.*method.*integration.*"):
+    with pytest.raises(
+        ValueError, match=r"'rk4.jit' integration method only supports deterministic integration.*"
+    ):
         config.set_file(f"{DATA_DIR}/config.yml")
 
         first_sim_index = 1
diff --git a/flepimop/gempyor_pkg/tests/statistics/test_statistic_class.py b/flepimop/gempyor_pkg/tests/statistics/test_statistic_class.py
index 4843986c1..bc2007270 100644
--- a/flepimop/gempyor_pkg/tests/statistics/test_statistic_class.py
+++ b/flepimop/gempyor_pkg/tests/statistics/test_statistic_class.py
@@ -8,6 +8,7 @@
 import pytest
 import scipy
 import xarray as xr
+import re
 
 from gempyor.statistics import Statistic
 from gempyor.testing import create_confuse_configview_from_dict
@@ -248,9 +249,7 @@ def test_unsupported_regularizations_value_error(
             ]
             if reg_name not in ["forecast", "allsubpop"]
         )
-        with pytest.raises(
-            ValueError, match=rf"^Unsupported regularization\: {unsupported_name}$"
-        ):
+        with pytest.raises(ValueError, match=rf"^Unsupported regularization \[received: 'invalid'\]"):
             mock_inputs.create_statistic_instance()
 
     @pytest.mark.parametrize("factory", all_valid_factories)
@@ -522,10 +521,10 @@ def test_compute_logloss_data_misshape_value_error(
 
         model_rows, model_cols = mock_inputs.model_data[mock_inputs.config["sim_var"]].shape
         gt_rows, gt_cols = mock_inputs.gt_data[mock_inputs.config["data_var"]].shape
-        expected_match = (
-            rf"^{mock_inputs.name} Statistic error\: data and groundtruth do not have "
-            rf"the same shape\: model\_data\.shape\=\({model_rows}\, {model_cols}\) "
-            rf"\!\= gt\_data\.shape\=\({gt_rows}\, {gt_cols}\)$"
+        expected_match = re.escape(
+            rf"`model_data` and `gt_data` do not have the same shape: "
+            rf"`model_data.shape` = '{mock_inputs.model_data[mock_inputs.config['sim_var']].shape}' "
+            rf"!= `gt_data.shape` = '{mock_inputs.gt_data[mock_inputs.config['data_var']].shape}'."
         )
         with pytest.raises(ValueError, match=expected_match):
             statistic.compute_logloss(mock_inputs.model_data, mock_inputs.gt_data)
diff --git a/flepimop/gempyor_pkg/tests/utils/test_random_distribution_sampler.py b/flepimop/gempyor_pkg/tests/utils/test_random_distribution_sampler.py
index 4ca9c0ac7..0dcd110fb 100644
--- a/flepimop/gempyor_pkg/tests/utils/test_random_distribution_sampler.py
+++ b/flepimop/gempyor_pkg/tests/utils/test_random_distribution_sampler.py
@@ -4,6 +4,7 @@
 
 import numpy as np
 import pytest
+import re
 
 from gempyor.testing import partials_are_similar
 from gempyor.utils import random_distribution_sampler
@@ -14,7 +15,7 @@ class TestRandomDistributionSampler:
     def test_not_implemented_error_exception(self, distribution: str) -> None:
         with pytest.raises(
             NotImplementedError,
-            match=rf"^unknown distribution \[got\: {distribution}\]$",
+            match=re.escape(rf"Unknown distribution [received: '{distribution}']."),
         ):
             random_distribution_sampler(distribution)
 
@@ -22,7 +23,7 @@ def test_not_implemented_error_exception(self, distribution: str) -> None:
     def test_binomial_p_value_error(self, p: float) -> None:
         with pytest.raises(
             ValueError,
-            match=rf"^p value {p} is out of range \[0\,1\]$",
+            match=re.escape(rf"Invalid `p-value`: '{p}' is out of range [0,1]."),
         ):
             random_distribution_sampler("binomial", n=100, p=p)
 
diff --git a/flepimop/gempyor_pkg/tests/utils/test_read_df.py b/flepimop/gempyor_pkg/tests/utils/test_read_df.py
index 48f05dbd5..b591aecd9 100644
--- a/flepimop/gempyor_pkg/tests/utils/test_read_df.py
+++ b/flepimop/gempyor_pkg/tests/utils/test_read_df.py
@@ -38,13 +38,13 @@ def test_raises_not_implemented_error(self) -> None:
         """
         with pytest.raises(
             expected_exception=NotImplementedError,
-            match="Invalid extension txt. Must be 'csv' or 'parquet'.",
+            match=r"Invalid extension provided: '.txt'. Supported extensions are `.csv` or `.parquet`.",
         ) as _:
             with NamedTemporaryFile(suffix=".txt") as temp_file:
                 read_df(fname=temp_file.name)
         with pytest.raises(
             expected_exception=NotImplementedError,
-            match="Invalid extension txt. Must be 'csv' or 'parquet'.",
+            match=r"Invalid extension provided: '.txt'. Supported extensions are `.csv` or `.parquet`.",
         ) as _:
             with NamedTemporaryFile(suffix=".txt") as temp_file:
                 fname = temp_file.name[:-4]
diff --git a/flepimop/gempyor_pkg/tests/utils/test_utils.py b/flepimop/gempyor_pkg/tests/utils/test_utils.py
index 87e7c4fbb..7b6606166 100644
--- a/flepimop/gempyor_pkg/tests/utils/test_utils.py
+++ b/flepimop/gempyor_pkg/tests/utils/test_utils.py
@@ -37,7 +37,8 @@ def test_read_df_and_write_success(fname, extension):
     ("fname", "extension"), [("mobility", "csv"), ("usa-geoid-params-output", "parquet")]
 )
 def test_read_df_and_write_fail(fname, extension):
-    with pytest.raises(NotImplementedError, match=r".*Invalid.*extension.*Must.*"):
+    with pytest.raises(
+        NotImplementedError, match=r"Invalid extension provided: '.'. Supported extensions are `.csv` or `.parquet`."):
         os.chdir(tmp_path)
         os.makedirs("data", exist_ok=True)
         os.chdir("data")
@@ -56,7 +57,7 @@ def test_read_df_and_write_fail(fname, extension):
 
 @pytest.mark.parametrize(("fname", "extension"), [("mobility", "")])
 def test_read_df_fail(fname, extension):
-    with pytest.raises(NotImplementedError, match=r".*Invalid.*extension.*"):
+    with pytest.raises(NotImplementedError, match=r"Invalid extension provided: '.'. Supported extensions are `.csv` or `.parquet`."):
         os.chdir(tmp_path)
         utils.read_df(fname=f"{DATA_DIR}/" + fname, extension=extension)
 
diff --git a/flepimop/gempyor_pkg/tests/utils/test_write_df.py b/flepimop/gempyor_pkg/tests/utils/test_write_df.py
index b13e0b948..e63afa2e5 100644
--- a/flepimop/gempyor_pkg/tests/utils/test_write_df.py
+++ b/flepimop/gempyor_pkg/tests/utils/test_write_df.py
@@ -30,13 +30,13 @@ def test_raises_not_implemented_error(self) -> None:
         """
         with pytest.raises(
             expected_exception=NotImplementedError,
-            match="Invalid extension txt. Must be 'csv' or 'parquet'.",
+            match=r"Invalid extension provided: '.txt'. Supported extensions are `.csv` or `.parquet`.",
         ) as _:
             with NamedTemporaryFile(suffix=".txt") as temp_file:
                 write_df(fname=temp_file.name, df=self.sample_df)
         with pytest.raises(
             expected_exception=NotImplementedError,
-            match="Invalid extension txt. Must be 'csv' or 'parquet'.",
+            match=r"Invalid extension provided: '.txt'. Supported extensions are `.csv` or `.parquet`.",
         ) as _:
             with NamedTemporaryFile(suffix=".txt") as temp_file:
                 fname = temp_file.name[:-4]